Hydrogen energy endurance supply control management system and method suitable for hydrogen-powered UAVs

Abstract

The present invention discloses a hydrogen energy endurance supply control management system and method suitable for hydrogen-powered drones, belonging to the field of energy supply control technology. The present invention constructs a theoretical endurance mileage model to calculate the theoretical endurance mileage that a drone can achieve under ideal conditions after using up a hydrogen storage bottle; collects the drone's flight data and inputs it into a pre-trained artificial intelligence algorithm model to obtain the drone's endurance loss in the current flight state, thereby obtaining a first endurance mileage; obtains the flight attitude combination used by the drone during flight and calculates the predicted endurance mileage; based on the predicted endurance mileage of the drone that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottle, screens out drones that are compatible with the hydrogen concentration of the remaining hydrogen storage bottle, and preferentially allocates the hydrogen concentration of the hydrogen storage bottle to the compatible drone; during the hydrogen refueling process, calculates the required hydrogen energy capacity and the corresponding hydrogen concentration.

Description

Technical Field

[0001] The present invention relates to the technical field of energy supply regulation and control, and in particular to a hydrogen energy endurance supply regulation and management system and method suitable for hydrogen-powered unmanned aerial vehicles. Background Art

[0002] With the continuous advancement of technology, drones are increasingly being used in various fields. However, drone endurance has always been a key factor limiting their wider application and greater effectiveness. Traditional lithium-battery-powered drones have relatively low energy density and limited range, making them difficult to meet the needs of long-distance and long-duration operations. Hydrogen energy, as a clean and efficient energy source with the advantage of high energy density, has gradually become a key research direction for improving drone endurance.

[0003] When estimating the range of a drone, existing technologies often only consider simple parameters such as battery capacity and motor power, without fully integrating the complex flight conditions and the characteristics of hydrogen energy for comprehensive calculations. During the flight of a drone, the flight speed, flight path, and flight attitude are constantly changing. Different flight attitudes will affect energy consumption. The concentration of the fuel cell on the drone will continue to change with each distance traveled. Existing technologies make it difficult to monitor and analyze the impact of these changes on energy consumption in real time. When refueling multiple drones, existing technologies lack consideration of the compatibility of the drone fuel cell with the hydrogen concentration of the hydrogen storage bottle. A simple average distribution or first-come, first-served method is usually used. This may result in some drones not being able to operate their fuel cells at their optimal state due to mismatched hydrogen concentrations. Summary of the Invention

[0004] The purpose of the present invention is to provide a hydrogen energy endurance supply control management system and method suitable for hydrogen-powered UAVs, so as to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a method for regulating and managing the supply of hydrogen energy for a hydrogen-powered UAV, comprising the following steps:

[0007] Collect the parameters of the drone and hydrogen storage bottle, build a theoretical range model, and calculate the theoretical range that the drone can achieve after using up a bottle of hydrogen storage bottle under ideal conditions;

[0008] Collect the drone's flight data, including flight speed, flight path, and flight attitude, and input it into a pre-trained artificial intelligence algorithm model to obtain the drone's endurance loss in the current flight state; deduct the endurance loss from the theoretical range to obtain the first range;

[0009] Based on the flight data of the UAV, the flight attitude combination used by the UAV during flight is obtained; when the UAV does not use the flight attitude combination, the first cruising range is used as the predicted cruising range; when the UAV uses the flight attitude combination, the saved mileage is calculated by analyzing the flight attitude changes and the corresponding energy savings; the saved mileage is accumulated to the first cruising range to serve as the predicted cruising range;

[0010] When the remaining hydrogen amount of the drone drops to the preset hydrogen refueling threshold, a hydrogen refueling demand signal is issued; based on the predicted cruising range of the drone that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottle, drones that are compatible with the hydrogen concentration of the remaining hydrogen storage bottle are screened out, and the hydrogen concentration of the hydrogen storage bottle is preferentially allocated to the compatible drones; during the hydrogen refueling process, the required hydrogen energy capacity and corresponding hydrogen concentration are obtained by calculating the amount of hydrogen energy required for hydrogen refueling of each drone and the currently allocated hydrogen energy.

[0011] In conjunction with the first aspect, in a first implementation of the first aspect of the present application, the collecting of drone parameters and hydrogen storage bottle parameters, constructing a theoretical cruising range model, and calculating the theoretical cruising range that the drone can achieve after using up a hydrogen storage bottle under ideal conditions includes:

[0012] The UAV parameters include aerodynamic parameters, power system parameters and flight performance parameters; the hydrogen storage bottle parameters include the volume and hydrogen concentration of the hydrogen storage bottle;

[0013] According to the volume V of the hydrogen storage bottle and the hydrogen concentration c, combined with the density of hydrogen , calculate the mass of hydrogen in the hydrogen storage bottle ; Using the heat of hydrogen combustion , and the total chemical energy of hydrogen is ; Obtain fuel cell power generation efficiency from power system parameters and the motor's drive efficiency , converting the total chemical energy of hydrogen into mechanical energy that can be used by the drone ;

[0014] The cruising speed of the UAV is obtained from the flight performance parameters, and the drag coefficient and wing area of the UAV are obtained from the aerodynamic parameters. According to the principles of aerodynamics, the drag D experienced by the UAV during flight is expressed as ,in, is the air density, is the cruising speed of the drone, is the drag coefficient of the drone, S is the wing area of the drone; the power required by the drone is ;

[0015] According to the law of conservation of energy, energy is equal to power multiplied by time, and the formula is , calculate the theoretical endurance of the drone , combined with the cruising speed of the drone, the theoretical cruising range is obtained .

[0016] In combination with the first aspect, in the second implementation of the first aspect of the present application, the flight data of the drone, including flight speed, flight path, and flight attitude, is collected and input into a pre-trained artificial intelligence algorithm model to obtain the endurance loss of the drone in the current flight state, including:

[0017] The artificial intelligence algorithm model selects a multi-layer perceptron model, divides the flight data of the drone into a training set, a validation set and a test set, and uses the training set data to train the model. During the training process, the model parameters are adjusted to minimize the error between the model's prediction results and the actual endurance loss, and the model parameters are updated using the gradient descent algorithm; the model is verified using the validation set and the model's hyperparameters are adjusted; the trained model is evaluated using the test set data, the model's prediction error is calculated, and the model is optimized.

[0018] In combination with the first aspect, in a third implementation of the first aspect of the present application, obtaining the flight attitude combination used by the drone during flight based on the flight data of the drone includes:

[0019] Based on the UAV's flight data, the UAV's attitude angles, including pitch, roll, and yaw angles, are calculated using a complementary filtering algorithm. Different attitude combinations are defined according to the UAV's flight mission. A clustering algorithm is used to analyze the UAV's attitude angles, identify different attitude combinations, and divide similar attitude data points into different clusters, with each cluster representing a posture combination.

[0020] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, the use of a clustering algorithm to analyze the attitude angles of the drone and identify different attitude combinations includes:

[0021] The attitude angles of the drones are integrated into a data set. The clustering algorithm selects the K-means clustering algorithm. The K value is determined by the elbow method. The sum of squared errors of clustering under different K values is calculated. Specifically, the sum of squared distances from each data point to the center of the cluster to which it belongs is calculated. K points are randomly selected in the data set as the initial cluster centers.

[0022] For each attitude angle data point in the dataset, use Euclidean distance to calculate its distance to the K cluster centers; assign each data point to the cluster with the cluster center closest to it; recalculate the mean of the data points in each cluster and use it as the new cluster center; repeat the above steps and iterate continuously until the cluster center no longer changes. At this time, the attitude angle data is divided into K different clusters.

[0023] In conjunction with the first aspect, in a fifth implementation of the first aspect of the present application, when the drone uses a flight attitude combination, the mileage saved is calculated by analyzing the flight attitude changes and the corresponding energy savings, including:

[0024] Based on the principles of flight mechanics and the power system parameters of the UAV, an energy consumption model is established, considering the effects of flight attitude, flight speed, flight altitude and air density on energy consumption. It is assumed that the UAV flies in a conventional flight attitude without using a flight attitude combination. According to the energy consumption model and combined with the power system parameters, the normal energy consumption under the same flight conditions is calculated. The actual energy consumption of the UAV when using the flight attitude combination is collected. The energy savings are calculated by comparing the normal energy consumption with the actual energy consumption. A relationship model between energy and flight mileage is established, and the energy savings are substituted into the relationship model to calculate the saved mileage.

[0025] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, the method of screening out drones that are compatible with the hydrogen concentration of the remaining hydrogen storage bottles based on the predicted cruising range of the drone that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottles, and preferentially allocating the hydrogen concentration of the hydrogen storage bottles to the compatible drones, includes:

[0026] Based on the operating characteristics of the fuel cell carried by each drone, its compatible range for hydrogen concentration is determined and an adaptation standard is formulated. When the hydrogen concentration of the hydrogen storage tank is within the compatible range for a particular drone, the drone is deemed compatible with the hydrogen concentration of the hydrogen storage tank.

[0027] Screen and adapt drones for hydrogen concentration distribution.

[0028] In combination with the first aspect, in a seventh implementation of the first aspect of the present application, the screening of suitable drones and the allocation of hydrogen concentration include:

[0029] Taking each hydrogen storage bottle as an object, we traverse all drones that need hydrogen refueling. For each hydrogen storage bottle, we check whether each drone meets its adaptive range for hydrogen concentration. We record the drones that meet the adaptive standards and form a list of compatible drones. Each list of compatible drones corresponds to a hydrogen storage bottle, and the list contains the number, model and predicted range of the compatible drone.

[0030] In each list of compatible drones, the drones are sorted according to their predicted range, with priority given to drones with longer predicted ranges; hydrogen concentration is allocated to the compatible drones from the hydrogen storage bottles in the determined allocation order, and the amount of hydrogen allocated each time is determined based on the design of the drone's fuel cell and the control accuracy of the hydrogen refueling equipment; the allocation operation is continued until the hydrogen capacity of the hydrogen storage bottle can no longer meet the hydrogen refueling needs of the next compatible drone, or all compatible drones of the hydrogen storage bottle have completed hydrogen refueling; the screening and allocation steps are repeated for the next hydrogen storage bottle until all drones that need hydrogen refueling have been processed or the hydrogen in all hydrogen storage bottles has been allocated.

[0031] In combination with the first aspect, in an eighth implementation of the first aspect of the present application, during the hydrogen refueling process, the required hydrogen energy capacity and the corresponding hydrogen concentration are obtained by calculating the amount of hydrogen energy required for hydrogen refueling of each drone and the currently allocated amount of hydrogen energy, including:

[0032] The total amount of electrical energy required by the drone during its planned flight is calculated in real time, and the amount of hydrogen energy required for hydrogen refueling is converted using the power generation efficiency. Each time the drone is refueled, the metering device of the hydrogen refueling equipment monitors the amount of hydrogen charged in real time, converts it into energy units, and records the number of the hydrogen storage bottle used for this refueling and the hydrogen concentration. As the hydrogen refueling process progresses, when a drone undergoes multiple refuelings, the amount of hydrogen energy used for each refueling is continuously accumulated.

[0033] At any time during the hydrogen refueling process, for all drones that need hydrogen refueling, the amount of hydrogen energy required for each drone is accumulated to obtain the total amount of hydrogen energy required; the hydrogen concentration of the current remaining hydrogen storage bottles is obtained in real time. When it is found that the existing hydrogen storage bottles cannot meet the needs of all un-hydrogenated drones, the system will issue an alarm to prompt the operator to allocate a new hydrogen source.

[0034] In a second aspect, the present invention provides a hydrogen energy endurance supply control and management system suitable for hydrogen-powered UAVs, comprising:

[0035] Theoretical cruising range calculation module: includes: a theoretical cruising range model construction unit and a theoretical cruising range calculation unit; wherein, the theoretical cruising range model construction unit collects UAV parameters and hydrogen storage bottle parameters to construct a theoretical cruising range model; the theoretical cruising range calculation unit calculates the theoretical cruising range that the UAV can achieve after using up a bottle of hydrogen storage bottle under ideal conditions;

[0036] The first cruising range calculation module includes: a flight data acquisition unit, an artificial intelligence algorithm processing unit, and a first cruising range calculation unit; wherein the flight data acquisition unit collects the UAV's flight data, including flight speed, flight path, and flight attitude, and the artificial intelligence algorithm processing unit inputs this data into a pre-trained artificial intelligence algorithm model to obtain the UAV's cruising range loss in the current flight state; the first cruising range calculation unit deducts the cruising range loss from the theoretical cruising range to obtain the first cruising range;

[0037] The predicted cruising range calculation module includes: a flight attitude combination recognition unit, an energy saving analysis unit, and a predicted cruising range calculation unit. The flight attitude combination recognition unit obtains the flight attitude combination used by the UAV during flight based on the UAV's flight data. The energy saving analysis unit uses the first cruising range as the predicted cruising range when the UAV does not use the flight attitude combination. When the UAV uses the flight attitude combination, it calculates the saved mileage by analyzing the flight attitude changes and the corresponding energy savings. The predicted cruising range calculation unit adds the saved mileage to the first cruising range as the predicted cruising range.

[0038] Hydrogen refueling demand and allocation module: includes: a hydrogen refueling demand monitoring unit, a hydrogen concentration adaptation screening and allocation unit, and a hydrogen refueling energy calculation unit; among them, the hydrogen refueling demand monitoring unit sends a hydrogen refueling demand signal when the remaining hydrogen amount of the UAV drops to a preset hydrogen refueling threshold; the hydrogen concentration adaptation screening and allocation unit screens out UAVs that are adapted to the hydrogen concentration of the remaining hydrogen storage bottles based on the predicted cruising range of the UAVs that need hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottles, and preferentially allocates the hydrogen concentration of the hydrogen storage bottles to the adapted UAVs; during the hydrogen refueling process, the hydrogen refueling energy calculation unit calculates the amount of hydrogen energy required for hydrogen refueling of each UAV and the currently allocated hydrogen energy to obtain the required hydrogen energy capacity and corresponding hydrogen concentration.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention collects UAV parameters and hydrogen storage tank parameters, combines the law of conservation of energy with the principles of flight mechanics to construct a theoretical range model, uses artificial intelligence algorithms to analyze flight data, accurately calculates the range loss under different flight states, and comprehensively considers the energy savings brought about by the combination of flight postures, thereby obtaining a highly accurate prediction of the range.

[0041] 2. The present invention uses artificial intelligence algorithms to analyze in real time the impact of the drone's flight speed, flight path, and flight attitude data on flight endurance loss. It can adjust the hydrogen energy supply strategy in a timely manner according to the real-time changes in the flight status, ensuring that the drone can operate efficiently in various complex flight scenarios.

[0042] 3. Based on the predicted cruising range of the drone and the hydrogen capacity and concentration of the hydrogen storage bottle, the present invention screens out drones that are compatible with the hydrogen concentration of the hydrogen storage bottle and prioritizes hydrogen concentration allocation, ensuring that the fuel cell of each drone operates under the optimal hydrogen concentration conditions, improving energy utilization efficiency, and extending the drone's cruising range. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a schematic diagram of the steps of the hydrogen energy endurance supply control management method applicable to hydrogen-powered UAVs of the present invention;

[0044] Figure 2 This is a system structure diagram of the hydrogen energy endurance supply control and management system applicable to hydrogen-powered UAVs according to the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution.

[0047] like Figure 1 The present invention is applicable to the hydrogen energy endurance supply regulation and management method of hydrogen-powered UAVs. The present invention provides a hydrogen energy endurance supply regulation and management method applicable to hydrogen-powered UAVs, including the following steps:

[0048] Step S100: Collect the parameters of the drone and the hydrogen storage bottle, build a theoretical cruising range model, and calculate the theoretical cruising range that the drone can achieve after using up a bottle of hydrogen storage bottle under ideal conditions;

[0049] Specifically, the UAV parameters include aerodynamic parameters, power system parameters and flight performance parameters; the hydrogen storage bottle parameters include the volume and hydrogen concentration of the hydrogen storage bottle;

[0050] According to the volume V of the hydrogen storage bottle and the hydrogen concentration c, combined with the density of hydrogen , calculate the mass of hydrogen in the hydrogen storage bottle ; Using the heat of hydrogen combustion , and the total chemical energy of hydrogen is ; Obtain fuel cell power generation efficiency from power system parameters and the motor's drive efficiency , converting the total chemical energy of hydrogen into mechanical energy that can be used by the drone ;

[0051] The cruising speed of the UAV is obtained from the flight performance parameters, and the drag coefficient and wing area of the UAV are obtained from the aerodynamic parameters. According to the principles of aerodynamics, the drag D experienced by the UAV during flight is expressed as ,in, is the air density, is the cruising speed of the drone, is the drag coefficient of the drone, S is the wing area of the drone; the power required by the drone is ;

[0052] According to the law of conservation of energy, energy is equal to power multiplied by time, and the formula is , calculate the theoretical endurance of the drone , combined with the cruising speed of the drone, the theoretical cruising range is obtained .

[0053] In one embodiment, the drag coefficient of the drone is =0.25, wing area S=1.5 square meters, fuel cell power generation efficiency =0.6, the driving efficiency of the motor =0.85. The cruising speed of the drone =20 m / s. The volume of the hydrogen storage bottle V = 5 liters, and the hydrogen concentration c = 99%. The density of hydrogen The heat of combustion of hydrogen is 0.0899 kg / m3. The density of air is 142,000 kJ / kg. It is 1.225 kg / cubic meter.

[0054] Convert the volume of the hydrogen storage bottle V = 5 liters to cubic meters, 5 liters = 0.005 cubic meters. Since the hydrogen concentration c = 99%, the actual volume of hydrogen is Cubic meter; According to the density formula, calculate the mass of hydrogen kilogram.

[0055] The total chemical energy is calculated using the heat of combustion of hydrogen. Kilojoules. Mechanical energy available to the drone Kilojoules.

[0056] According to the principles of aerodynamics, the resistance a drone encounters during flight Awesome. The power required by the drone is P = 91.875 * 20 = 1837.5 watts.

[0057] According to the law of conservation of energy, t = 32280 / 1837.5 ≈ 17.57 seconds. Combined with the drone's cruising speed, the theoretical range is calculated as L = 20 * 17.57 = 351.4 meters.

[0058] Step S200: collecting the flight data of the UAV, including the flight speed, flight path, and flight attitude, and inputting it into a pre-trained artificial intelligence algorithm model to obtain the endurance loss of the UAV in the current flight state; deducting the endurance loss from the theoretical endurance mileage to obtain a first endurance mileage;

[0059] Specifically, the artificial intelligence algorithm model selects a multi-layer perceptron model, divides the flight data of the drone into a training set, a validation set, and a test set, and uses the training set data to train the model. During the training process, by adjusting the parameters of the model, the error between the model's prediction results and the actual endurance loss is minimized, and the model parameters are updated using the gradient descent algorithm; the model is verified using the validation set and the model's hyperparameters are adjusted; the trained model is evaluated using the test set data, the model's prediction error is calculated, and the model is optimized.

[0060] In one specific embodiment, the collected flight speed, flight path (latitude and longitude coordinates converted into displacement vectors relative to the starting point), and flight attitude data are integrated into a feature matrix, with each row representing the flight status data at a specific time point. Simultaneously, the corresponding actual flight endurance loss is organized into a label vector. The data is divided into training, validation, and test sets in a ratio of 70%, 15%, and 15%. That is, from 6,000 data points, 4,200 are selected as the training set, 900 as the validation set, and 900 as the test set. This ensures that the data in each set represents different flight phases and states.

[0061] Build a multilayer perceptron model with two hidden layers. The number of nodes in the input layer is determined by the number of features. Since there are six features in total: flight speed, flight path (represented by two displacement vectors), and flight attitude (three angles), the input layer has six nodes. The first hidden layer has 10 nodes, the second hidden layer has 8 nodes, and the output layer has one node, which is used to predict the range loss. ReLU is selected as the activation function for the hidden layer. No activation function is used in the output layer because range loss is a continuous value.

[0062] The model was trained using the training set data for 100 epochs. In each epoch, the model's predictions were calculated using forward propagation. The mean squared error (MSE) was then used as the loss function to calculate the difference between the predicted results and the actual range loss. Model parameters were updated using the gradient descent algorithm with a learning rate of 0.001. When updating parameters, the gradient of the loss function with respect to the parameters was used to adjust the parameter values in the direction that reduced the loss function. During training, the model was validated using the validation set data every five epochs. The validation set data was fed into the model, and the model loss on the validation set was calculated (also using the MSE). Based on changes in the validation set loss, the model's hyperparameters were adjusted. For example, if the validation set loss stopped decreasing or even increased over several consecutive epochs, hyperparameters such as the number of hidden layer nodes and the learning rate were adjusted. Adjusting the learning rate to 0.0005 reduced the validation set loss, indicating that the adjustments were effective. The model achieved a mean squared error of 25 and a mean absolute error of 3 kJ on the test set. The number of hidden layers was increased. After adding a hidden layer and retraining, it was found that the mean square error of the model on the test set was reduced to 20, and the mean absolute error was reduced to 2.5 kilojoules, indicating that the model performance was improved.

[0063] The flight data collected in real time is input into the optimized model to obtain the predicted range loss. Under the current flight state, the model predicts a range loss of 100 kilojoules. Knowing that the energy corresponding to the previously calculated theoretical range is 32.28 kilojoules, the predicted range loss is deducted from the energy corresponding to the theoretical range. Since there is a certain conversion relationship between energy and mileage (based on previously calculated parameters such as power and speed), the first range is calculated through conversion. The conversion relationship is that every kilojoule of energy corresponds to 10 meters of mileage, so the first range = (32.28-10)*10=222.8 meters.

[0064] Step S300: Based on the flight data of the UAV, a flight attitude combination used by the UAV during flight is obtained; when the UAV does not use the flight attitude combination, the first cruising range is used as the predicted cruising range; when the UAV uses the flight attitude combination, the saved mileage is calculated by analyzing the flight attitude changes and the corresponding energy savings; the saved mileage is added to the first cruising range to serve as the predicted cruising range;

[0065] Specifically, based on the UAV's flight data, the UAV's attitude angles, including pitch angle, roll angle and yaw angle, are calculated through a complementary filtering algorithm; different attitude combinations are defined according to the UAV's flight mission; a clustering algorithm is used to analyze the UAV's attitude angles, identify different attitude combinations, and divide similar attitude data points into different clusters, with each cluster representing a posture combination.

[0066] Furthermore, the attitude angles of the drones are integrated into a data set. The clustering algorithm selects the K-means clustering algorithm, and the K value is determined by the elbow method. The sum of squared errors of clustering under different K values is calculated, specifically the sum of squared distances from each data point to the center of the cluster to which it belongs. K points are randomly selected in the data set as the initial cluster centers.

[0067] For each attitude angle data point in the dataset, use Euclidean distance to calculate its distance to the K cluster centers; assign each data point to the cluster with the cluster center closest to it; recalculate the mean of the data points in each cluster and use it as the new cluster center; repeat the above steps and iterate continuously until the cluster center no longer changes. At this time, the attitude angle data is divided into K different clusters.

[0068] Furthermore, based on the principles of flight mechanics and the power system parameters of the UAV, an energy consumption model is established by considering the effects of flight attitude, flight speed, flight altitude and air density on energy consumption; it is assumed that the UAV flies according to a conventional flight attitude without using a flight attitude combination, and according to the energy consumption model and combined with the power system parameters, the normal energy consumption under the same flight conditions is calculated; the actual energy consumption of the UAV when using the flight attitude combination is collected; the energy savings are calculated by comparing the normal energy consumption and the actual energy consumption; a relationship model between energy and flight mileage is established, and the energy savings are substituted into the relationship model to calculate the saved mileage.

[0069] In a specific embodiment, based on the logistics delivery flight mission performed by the drone, the following posture combinations are defined:

[0070] Level cruise attitude combination: pitch angle between −5° and 5°, roll angle between −3° and 3°, and yaw angle change rate is small (no more than 5° per minute).

[0071] Turn attitude combination: The yaw angle changes at a large rate (more than 10° per minute), and the pitch and roll angles are adjusted accordingly based on the turn direction. For example, when turning left, the roll angle is positive, and when turning right, the roll angle is negative. The pitch angle is adjusted between -10° and 10° to maintain flight balance.

[0072] Climb attitude combination: the pitch angle is greater than 10°, the roll angle is between −5° and 5°, and the yaw angle remains relatively stable.

[0073] The calculated attitude angle data of 3000 time points are integrated into a data set, and each data point contains the pitch angle, roll angle and yaw angle at the corresponding moment.

[0074] The K value is determined by the elbow method. Tests are performed in the range of K=2 to K=10. For each K value, the K-means clustering algorithm is executed. For example, when K=3, 3 points are randomly selected in the data set as the initial cluster centers C1, C2, and C3. The sum of squared errors (SSE) of the clustering under different K values is calculated, that is, the sum of squared distances from each data point to the center of the cluster to which it belongs. Each data point is assigned to the cluster with the closest cluster center. Then the mean of the data points in each cluster is recalculated and used as the new cluster center. Repeat this process until the cluster center no longer changes. Record the SSE value at this time. By plotting the relationship between the K value and the SSE value, it is found that when K=3, the curve has an obvious elbow point, so K=3 is determined to be the appropriate number of clusters.

[0075] Clustering was repeated with the fixed K=3. Three initial cluster centers were randomly selected again, and the above steps of distance calculation, data point assignment, and cluster center update were repeated. After multiple iterations, the attitude angle dataset was divided into three different clusters. Each cluster was analyzed, and the attitude angle range and variation characteristics of the data points in the cluster were matched with predefined attitude combinations. It was found that the attitude angle range of the data points in Cluster 1 matched the definition of the level cruise attitude combination, Cluster 2 matched the turning attitude combination, and Cluster 3 matched the climbing attitude combination.

[0076] Based on the principles of flight mechanics, the energy consumption model is established by considering the effects of flight attitude, flight speed, flight altitude and air density on energy consumption. Assume that the air density is , the flight speed is v, the wing area is S, and the drag coefficient is , the lift coefficient is , the acceleration due to gravity is g, and the mass of the drone is m.

[0077] Flight resistance , lift , in the level cruising attitude, and is at a relatively stable design value; in order to maintain the centripetal force in the turning posture, and It will be adjusted according to the turning radius and speed. In terms of power system parameters, it is known that the fuel cell power generation efficiency and motor drive efficiency , the power required for drone flight Based on these parameters, an energy consumption model E=P*t is established, where t is the flight time. In this embodiment, the normal energy consumption is calculated to be 1058820 joules.

[0078] During this flight segment, actual power output varied over time. Integrating this energy consumption yielded an actual energy consumption of 900,000 joules. The energy savings were calculated as 1,058,820 - 900,000 = 158,820 joules. A relationship model between energy consumption and flight distance was established, and the energy savings were substituted into the model to calculate the distance saved: 158,820 / 1,000 = 158.82 meters.

[0079] Given that the first cruising range calculated previously was 222.8 meters, add the saved mileage to the first cruising range to get a predicted cruising range of 222.8 + 158.82 = 381.62 meters.

[0080] Step S400: When the remaining hydrogen amount of the UAV drops to a preset hydrogen refueling threshold, a hydrogen refueling demand signal is issued; based on the predicted cruising range of the UAV that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottle, the UAVs that are compatible with the hydrogen concentration of the remaining hydrogen storage bottle are screened out, and the hydrogen concentration of the hydrogen storage bottle is preferentially allocated to the compatible UAVs; during the hydrogen refueling process, the required hydrogen energy capacity and corresponding hydrogen concentration are obtained by calculating the amount of hydrogen energy required for hydrogen refueling of each UAV and the currently allocated hydrogen energy.

[0081] Specifically, based on the operating characteristics of the fuel cell carried by each drone, its adaptive range for hydrogen concentration is determined and an adaptation standard is formulated; when the hydrogen concentration of the hydrogen storage tank is within the adaptive range of a certain drone, the drone is determined to be compatible with the hydrogen concentration of the hydrogen storage tank;

[0082] Screen and adapt drones for hydrogen concentration distribution.

[0083] Furthermore, taking each hydrogen storage bottle as an object, we traverse all drones that need hydrogen refueling. For each hydrogen storage bottle, we check whether each drone meets its adaptation range for hydrogen concentration. The drones that meet the adaptation criteria are recorded to form a list of compatible drones. Each list of compatible drones corresponds to a hydrogen storage bottle, and the list contains the number, model and predicted range of the compatible drone.

[0084] In each list of compatible drones, the drones are sorted according to their predicted range, with priority given to drones with longer predicted ranges; hydrogen concentration is allocated to the compatible drones from the hydrogen storage bottles in the determined allocation order, and the amount of hydrogen allocated each time is determined based on the design of the drone's fuel cell and the control accuracy of the hydrogen refueling equipment; the allocation operation is continued until the hydrogen capacity of the hydrogen storage bottle can no longer meet the hydrogen refueling needs of the next compatible drone, or all compatible drones of the hydrogen storage bottle have completed hydrogen refueling; the screening and allocation steps are repeated for the next hydrogen storage bottle until all drones that need hydrogen refueling have been processed or the hydrogen in all hydrogen storage bottles has been allocated.

[0085] Furthermore, the total amount of electrical energy required by the drone during the planned flight is calculated in real time, and the amount of hydrogen energy is converted using the power generation efficiency to calculate the amount of hydrogen energy required for hydrogen refueling of the drone. Each time the drone is refueled, the metering device of the hydrogen refueling equipment monitors the amount of hydrogen charged in real time, converts it into energy units, and records the number of the hydrogen storage bottle used for this refueling and the hydrogen concentration. As the hydrogen refueling process progresses, when a drone undergoes multiple hydrogen refuelings, the amount of hydrogen energy for each refueling is continuously accumulated.

[0086] At any time during the hydrogen refueling process, for all drones that need hydrogen refueling, the amount of hydrogen energy required for each drone is accumulated to obtain the total amount of hydrogen energy required; the hydrogen concentration of the current remaining hydrogen storage bottles is obtained in real time. When it is found that the existing hydrogen storage bottles cannot meet the needs of all un-hydrogenated drones, the system will issue an alarm to prompt the operator to allocate a new hydrogen source.

[0087] In a specific embodiment, there are five drones that need hydrogen refueling, numbered U1, U2, U3, U4, and U5, and modeled Model A, Model B, Model A, Model C, and Model B. The fuel cell parameters of each drone are different. After preliminary testing and analysis, the adaptive range of their hydrogen concentrations is determined as follows:

[0088] U1 (Model A): Suitable for hydrogen concentration range of 95%-99%.

[0089] U2 (Model B): Suitable for hydrogen concentration range of 90%-96%.

[0090] U3 (Model A): Suitable for hydrogen concentration range of 95%-99%.

[0091] U4 (Model C): Suitable for hydrogen concentration range of 93%-97%.

[0092] U5 (Model B): Suitable for hydrogen concentration range of 90%-96%.

[0093] It is known that their predicted ranges are 300 kilometers, 250 kilometers, 320 kilometers, 280 kilometers and 260 kilometers respectively.

[0094] There are three remaining hydrogen storage bottles, numbered H1, H2, and H3. Their hydrogen capacity and concentration information are as follows:

[0095] H1: Hydrogen capacity is 500 liters and hydrogen concentration is 96%.

[0096] H2: Hydrogen capacity is 400 liters and hydrogen concentration is 93%.

[0097] H3: Hydrogen capacity is 350 liters and hydrogen concentration is 98%.

[0098] Each drone is equipped with a hydrogen level monitoring sensor that monitors the remaining hydrogen level in real time. The preset hydrogen refueling threshold is 20% of the total capacity. When the remaining hydrogen level of U1 drops to this threshold, it triggers a hydrogen refueling request signal. Subsequently, U2, U3, and other drones also reach the threshold and send signals.

[0099] Taking the H1 hydrogen storage bottle (hydrogen concentration 96%) as an example, we traverse 5 drones that need hydrogen refueling:

[0100] U1 (Model A), the adaptation range is 95%-99%, 96% is within its adaptation range, and the list of compatible drones from U1 to H1 is recorded.

[0101] U2 (Model B), adaptation range 90%-96%, 96% within its adaptation range, record the list of compatible drones from U2 to H1.

[0102] U3 (Model A), the adaptation range is 95%-99%, 96% is within its adaptation range, and the list of compatible drones from U3 to H1 is recorded.

[0103] U4 (Model C), adaptation range is 93%-97%, 96% is within its adaptation range, record the list of compatible drones from U4 to H1.

[0104] U5 (Model B), adaptation range is 90%-96%, 96% is within its adaptation range, record the list of compatible drones from U5 to H1.

[0105] The list of compatible drones for the H1 is sorted from highest to lowest predicted range: U3 (320 km), U1 (300 km), U4 (280 km), U5 (260 km), and U2 (250 km). Similarly, the list of compatible drones for the H2 and H3 hydrogen storage tanks is sorted and sorted. For the H2 (93% hydrogen concentration), the compatible drones are the U2, U4, and U5, and the sorting order is U4 (280 km), U5 (260 km), and U2 (250 km). For the H3 (98% hydrogen concentration), the compatible drones are the U1 and U3, and the sorting order is U3 (320 km) and U1 (300 km).

[0106] According to the order, U3 is refueled first. The amount of hydrogen to be dispensed at each time is determined based on the design of U3's fuel cell and the control accuracy of the refueling equipment. Calculations indicate that 150 liters of hydrogen are required for refueling U3. The refueling equipment begins operation, and the metering device monitors the amount of hydrogen being injected in real time. After refueling is complete, the hydrogen storage tank number H1 and the hydrogen concentration of 96% are recorded. At this point, the remaining hydrogen capacity of H1 is 500-150 = 350 liters. Next, U1 is refueled, requiring 120 liters of hydrogen. After refueling is complete, the remaining hydrogen capacity of H1 is 350-120 = 230 liters. U4 is refueled, requiring 100 liters of hydrogen. At this point, the remaining hydrogen capacity of H1 is 230-100 = 130 liters. When refueling U5, U5 requires 120 liters of hydrogen, but H1 has insufficient remaining hydrogen capacity, so refueling to H1 is stopped.

[0107] Then H2 and H3 are allocated.

[0108] Real-time information on the hydrogen concentration and capacity of the remaining hydrogen tanks is obtained. If the system detects that the available hydrogen tanks cannot meet the needs of all unrefueled drones, an alarm will be issued. If, based on the current allocation, the hydrogen energy in all remaining tanks is insufficient to meet the remaining hydrogen needs of the U2, the system will immediately issue an alarm, prompting the operator to deploy a new hydrogen source.

[0109] In a second aspect, the present invention provides a hydrogen energy endurance supply control and management system suitable for hydrogen-powered UAVs, comprising:

[0110] Theoretical cruising range calculation module: includes: a theoretical cruising range model construction unit and a theoretical cruising range calculation unit; wherein, the theoretical cruising range model construction unit collects UAV parameters and hydrogen storage bottle parameters to construct a theoretical cruising range model; the theoretical cruising range calculation unit calculates the theoretical cruising range that the UAV can achieve after using up a bottle of hydrogen storage bottle under ideal conditions;

[0111] The first cruising range calculation module includes: a flight data acquisition unit, an artificial intelligence algorithm processing unit, and a first cruising range calculation unit; wherein the flight data acquisition unit collects the UAV's flight data, including flight speed, flight path, and flight attitude, and the artificial intelligence algorithm processing unit inputs this data into a pre-trained artificial intelligence algorithm model to obtain the UAV's cruising range loss in the current flight state; the first cruising range calculation unit deducts the cruising range loss from the theoretical cruising range to obtain the first cruising range;

[0112] The predicted cruising range calculation module includes: a flight attitude combination recognition unit, an energy saving analysis unit, and a predicted cruising range calculation unit. The flight attitude combination recognition unit obtains the flight attitude combination used by the UAV during flight based on the UAV's flight data. The energy saving analysis unit uses the first cruising range as the predicted cruising range when the UAV does not use the flight attitude combination. When the UAV uses the flight attitude combination, it calculates the saved mileage by analyzing the flight attitude changes and the corresponding energy savings. The predicted cruising range calculation unit adds the saved mileage to the first cruising range as the predicted cruising range.

[0113] Hydrogen refueling demand and allocation module: includes: a hydrogen refueling demand monitoring unit, a hydrogen concentration adaptation screening and allocation unit, and a hydrogen refueling energy calculation unit; among them, the hydrogen refueling demand monitoring unit sends a hydrogen refueling demand signal when the remaining hydrogen amount of the UAV drops to a preset hydrogen refueling threshold; the hydrogen concentration adaptation screening and allocation unit screens out UAVs that are adapted to the hydrogen concentration of the remaining hydrogen storage bottles based on the predicted cruising range of the UAVs that need hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottles, and preferentially allocates the hydrogen concentration of the hydrogen storage bottles to the adapted UAVs; during the hydrogen refueling process, the hydrogen refueling energy calculation unit calculates the amount of hydrogen energy required for hydrogen refueling of each UAV and the currently allocated hydrogen energy to obtain the required hydrogen energy capacity and corresponding hydrogen concentration.

[0114] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A hydrogen energy endurance supply control and management method applicable to hydrogen-powered UAVs, characterized in that: The following steps are involved: Collect the parameters of the drone and hydrogen storage bottle, build a theoretical range model, and calculate the theoretical range that the drone can achieve after using up a bottle of hydrogen storage bottle under ideal conditions; Collect the drone's flight data, including flight speed, flight path, and flight attitude, and input it into a pre-trained artificial intelligence algorithm model to obtain the drone's endurance loss in the current flight state; deduct the endurance loss from the theoretical range to obtain the first range; Based on the flight data of the UAV, the flight attitude combination used by the UAV during flight is obtained; when the UAV does not use the flight attitude combination, the first cruising range is used as the predicted cruising range; when the UAV uses the flight attitude combination, the saved mileage is calculated by analyzing the flight attitude changes and the corresponding energy savings; the saved mileage is accumulated to the first cruising range to serve as the predicted cruising range; Based on the UAV's flight data, the complementary filtering algorithm is used to calculate the UAV's attitude angles, including pitch, roll, and yaw. Different attitude combinations are defined based on the UAV's flight mission. A clustering algorithm is used to analyze the UAV's attitude angles, identify different attitude combinations, and divide similar attitude data points into different clusters, with each cluster representing a different attitude combination. Based on flight mechanics principles and the UAV's power system parameters, an energy consumption model was established, taking into account the effects of flight attitude, flight speed, flight altitude, and air density on energy consumption. The UAV was assumed to fly in a conventional flight attitude without using any flight attitude combination. Based on the energy consumption model and the power system parameters, the normal energy consumption under the same flight conditions was calculated. The actual energy consumption of the UAV when using the flight attitude combination was collected. Calculate energy savings by comparing normal energy consumption with actual energy consumption; Establish a relationship model between energy and flight mileage, substitute the energy savings into the relationship model, and calculate the mileage saved; When the remaining hydrogen amount of the drone drops to the preset hydrogen refueling threshold, a hydrogen refueling demand signal is issued; based on the predicted cruising range of the drone that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottle, drones that are compatible with the hydrogen concentration of the remaining hydrogen storage bottle are screened out, and the hydrogen concentration of the hydrogen storage bottle is preferentially allocated to the compatible drones; during the hydrogen refueling process, the required hydrogen energy capacity and corresponding hydrogen concentration are obtained by calculating the amount of hydrogen energy required for hydrogen refueling of each drone and the currently allocated hydrogen energy.

2. The hydrogen energy endurance supply control and management method for hydrogen-powered UAVs according to claim 1 is characterized in that: The method of collecting the parameters of the drone and the hydrogen storage bottle, constructing a theoretical cruising range model, and calculating the theoretical cruising range that the drone can achieve after using up a bottle of hydrogen storage bottle under ideal conditions includes: The UAV parameters include aerodynamic parameters, power system parameters and flight performance parameters; the hydrogen storage bottle parameters include the volume and hydrogen concentration of the hydrogen storage bottle; According to the volume V of the hydrogen storage bottle and the hydrogen concentration c, combined with the density of hydrogen , calculate the mass of hydrogen in the hydrogen storage bottle ; Using the heat of hydrogen combustion , and the total chemical energy of hydrogen is ; Obtain fuel cell power generation efficiency from power system parameters and the motor's drive efficiency , converting the total chemical energy of hydrogen into mechanical energy that can be used by the drone ; The cruising speed of the UAV is obtained from the flight performance parameters, and the drag coefficient and wing area of the UAV are obtained from the aerodynamic parameters. According to the principles of aerodynamics, the drag D experienced by the UAV during flight is expressed as ,in, is the air density, is the cruising speed of the drone, is the drag coefficient of the drone, S is the wing area of the drone; the power required by the drone is ; According to the law of conservation of energy, energy is equal to power multiplied by time, and the formula is , calculate the theoretical endurance of the drone , combined with the cruising speed of the drone, the theoretical cruising range is obtained .

3. The hydrogen energy endurance supply control and management method for hydrogen-powered UAVs according to claim 1 is characterized in that: The flight data of the drone, including flight speed, flight path, and flight attitude, is collected and input into a pre-trained artificial intelligence algorithm model to obtain the endurance loss of the drone in the current flight state, including: The artificial intelligence algorithm model selects a multi-layer perceptron model, divides the flight data of the drone into a training set, a validation set and a test set, and uses the training set data to train the model. During the training process, the model parameters are adjusted to minimize the error between the model's prediction results and the actual endurance loss, and the model parameters are updated using the gradient descent algorithm; the model is verified using the validation set and the model's hyperparameters are adjusted; the trained model is evaluated using the test set data, the model's prediction error is calculated, and the model is optimized.

4. The hydrogen energy endurance supply control and management method for a hydrogen-powered UAV according to claim 1, wherein the clustering algorithm is used to analyze the attitude angles of the UAV and identify different attitude combinations, including: The attitude angles of the drones are integrated into a data set. The clustering algorithm selects the K-means clustering algorithm. The K value is determined by the elbow method. The sum of squared errors of clustering under different K values is calculated. Specifically, the sum of squared distances from each data point to the center of the cluster to which it belongs is calculated. K points are randomly selected in the data set as the initial cluster centers. For each attitude angle data point in the dataset, use Euclidean distance to calculate its distance to the K cluster centers; assign each data point to the cluster with the cluster center closest to it; recalculate the mean of the data points in each cluster and use it as the new cluster center; repeat the above steps and iterate continuously until the cluster center no longer changes. At this time, the attitude angle data is divided into K different clusters.

5. The hydrogen energy endurance supply control and management method for hydrogen-powered UAVs according to claim 1 is characterized in that: The method of screening out drones that are compatible with the hydrogen concentration of the remaining hydrogen storage bottles based on the predicted cruising range of the drone that needs hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottles, and preferentially allocating the hydrogen concentration of the hydrogen storage bottles to the compatible drones, includes: Based on the operating characteristics of the fuel cell carried by each drone, its compatible range for hydrogen concentration is determined and an adaptation standard is formulated. When the hydrogen concentration of the remaining hydrogen storage tank is within the compatible range for a particular drone, the drone is determined to be compatible with the hydrogen concentration of the hydrogen storage tank. Screen and adapt drones for hydrogen concentration distribution.

6. The hydrogen energy endurance supply control and management method for hydrogen-powered UAVs according to claim 5 is characterized in that: The screening of suitable drones and the distribution of hydrogen concentration include: Taking each hydrogen storage bottle as an object, we traverse all drones that need hydrogen refueling. For each hydrogen storage bottle, we check whether each drone meets its adaptive range for hydrogen concentration. We record the drones that meet the adaptive standards and form a list of compatible drones. Each list of compatible drones corresponds to a hydrogen storage bottle, and the list contains the number, model and predicted range of the compatible drone. In each list of compatible drones, the drones are sorted according to their predicted range, with priority given to drones with longer predicted ranges; hydrogen concentration is allocated to the compatible drones from the hydrogen storage bottles in the determined allocation order, and the amount of hydrogen allocated each time is determined based on the design of the drone's fuel cell and the control accuracy of the hydrogen refueling equipment; the allocation operation is continued until the hydrogen capacity of the hydrogen storage bottle can no longer meet the hydrogen refueling needs of the next compatible drone, or all compatible drones of the hydrogen storage bottle have completed hydrogen refueling; the screening and allocation steps are repeated for the next hydrogen storage bottle until all drones that need hydrogen refueling have been processed or the hydrogen in all hydrogen storage bottles has been allocated.

7. The hydrogen energy endurance supply control and management method for hydrogen-powered UAVs according to claim 1 is characterized in that: During the hydrogen refueling process, the required hydrogen energy capacity and the corresponding hydrogen concentration are obtained by calculating the amount of hydrogen energy required for each drone and the currently allocated hydrogen energy, including: The total amount of electrical energy required by the drone during its planned flight is calculated in real time, and the amount of hydrogen energy required for hydrogen refueling is converted using the power generation efficiency. Each time the drone is refueled, the metering device of the hydrogen refueling equipment monitors the amount of hydrogen charged in real time, converts it into energy units, and records the number of the hydrogen storage bottle used for this refueling and the hydrogen concentration. As the hydrogen refueling process progresses, when a drone undergoes multiple refuelings, the amount of hydrogen energy used for each refueling is continuously accumulated. At any time during the hydrogen refueling process, for all drones that need hydrogen refueling, the amount of hydrogen energy required for each drone is accumulated to obtain the total amount of hydrogen energy required; the hydrogen concentration of the current remaining hydrogen storage bottles is obtained in real time. When it is found that the existing hydrogen storage bottles cannot meet the needs of all un-hydrogenated drones, the system will issue an alarm to prompt the operator to allocate a new hydrogen source.

8. A hydrogen energy endurance supply control and management system applicable to hydrogen-powered UAVs, using the hydrogen energy endurance supply control and management method applicable to hydrogen-powered UAVs according to any one of claims 1 to 7, characterized in that: include: Theoretical cruising range calculation module: includes: a theoretical cruising range model construction unit and a theoretical cruising range calculation unit; wherein, the theoretical cruising range model construction unit collects UAV parameters and hydrogen storage bottle parameters to construct a theoretical cruising range model; the theoretical cruising range calculation unit calculates the theoretical cruising range that the UAV can achieve after using up a bottle of hydrogen storage bottle under ideal conditions; The first cruising range calculation module includes: a flight data acquisition unit, an artificial intelligence algorithm processing unit, and a first cruising range calculation unit; wherein the flight data acquisition unit collects the UAV's flight data, including flight speed, flight path, and flight attitude, and the artificial intelligence algorithm processing unit inputs this data into a pre-trained artificial intelligence algorithm model to obtain the UAV's cruising range loss in the current flight state; the first cruising range calculation unit deducts the cruising range loss from the theoretical cruising range to obtain the first cruising range; The predicted cruising range calculation module includes: a flight attitude combination recognition unit, an energy saving analysis unit, and a predicted cruising range calculation unit. The flight attitude combination recognition unit obtains the flight attitude combination used by the UAV during flight based on the UAV's flight data. The energy saving analysis unit uses the first cruising range as the predicted cruising range when the UAV does not use the flight attitude combination. When the UAV uses the flight attitude combination, it calculates the saved mileage by analyzing the flight attitude changes and the corresponding energy savings. The predicted cruising range calculation unit adds the saved mileage to the first cruising range as the predicted cruising range. Hydrogen refueling demand and allocation module: includes: a hydrogen refueling demand monitoring unit, a hydrogen concentration adaptation screening and allocation unit, and a hydrogen refueling energy calculation unit; among them, the hydrogen refueling demand monitoring unit sends a hydrogen refueling demand signal when the remaining hydrogen amount of the UAV drops to a preset hydrogen refueling threshold; the hydrogen concentration adaptation screening and allocation unit screens out UAVs that are adapted to the hydrogen concentration of the remaining hydrogen storage bottles based on the predicted cruising range of the UAVs that need hydrogen refueling and the hydrogen capacity and hydrogen concentration of the current remaining hydrogen storage bottles, and preferentially allocates the hydrogen concentration of the hydrogen storage bottles to the adapted UAVs; during the hydrogen refueling process, the hydrogen refueling energy calculation unit calculates the amount of hydrogen energy required for hydrogen refueling of each UAV and the currently allocated hydrogen energy to obtain the required hydrogen energy capacity and corresponding hydrogen concentration.

Images